DREAM-R: Multimodal Speculative Reasoning with RL-Based Refined Drafting, Precise Verification, and Fully Parallel Execution
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917540650811392 |
|---|---|
| author | Hu, Yunhai Liu, Zining Yin, Xiangyang Xia, Tianhua Bao, Bo Sather, Eric Thangarasa, Vithursan Zhang, Sai Qian |
| author_facet | Hu, Yunhai Liu, Zining Yin, Xiangyang Xia, Tianhua Bao, Bo Sather, Eric Thangarasa, Vithursan Zhang, Sai Qian |
| contents | Speculative reasoning has recently been proposed as a means to accelerate reasoning-intensive generation in large multimodal models, but its effectiveness is often constrained by misalignment between speculative drafts and target-verified reasoning. In this work, we introduce DREAM-R, a framework that substantially improves the performance of speculative reasoning. At its core, DREAM-R employs Speculative Alignment Policy Optimization (SAPO), a reinforcement-learning objective that trains draft models to generate reasoning steps that are both faithful to target trajectories and concise. We further propose a Threshold-based Verification Mechanism (TBVM) that uses a ratio-based criterion to provide stable and interpretable acceptance of speculative steps only when positive evidence clearly dominates, thereby preventing error propagation. Building on these components, we develop a Fully Parallel Speculative Reasoning (FPSR) framework that parallelizes draft generation, target-side reasoning, and verification across multi-step reasoning, enabling early stopping and clean fallback. Experiments on reasoning-heavy benchmarks demonstrate up to speedup while preserving target-model accuracy, yielding substantial efficiency gains without compromising reasoning quality. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_28678 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DREAM-R: Multimodal Speculative Reasoning with RL-Based Refined Drafting, Precise Verification, and Fully Parallel Execution Hu, Yunhai Liu, Zining Yin, Xiangyang Xia, Tianhua Bao, Bo Sather, Eric Thangarasa, Vithursan Zhang, Sai Qian Artificial Intelligence Speculative reasoning has recently been proposed as a means to accelerate reasoning-intensive generation in large multimodal models, but its effectiveness is often constrained by misalignment between speculative drafts and target-verified reasoning. In this work, we introduce DREAM-R, a framework that substantially improves the performance of speculative reasoning. At its core, DREAM-R employs Speculative Alignment Policy Optimization (SAPO), a reinforcement-learning objective that trains draft models to generate reasoning steps that are both faithful to target trajectories and concise. We further propose a Threshold-based Verification Mechanism (TBVM) that uses a ratio-based criterion to provide stable and interpretable acceptance of speculative steps only when positive evidence clearly dominates, thereby preventing error propagation. Building on these components, we develop a Fully Parallel Speculative Reasoning (FPSR) framework that parallelizes draft generation, target-side reasoning, and verification across multi-step reasoning, enabling early stopping and clean fallback. Experiments on reasoning-heavy benchmarks demonstrate up to speedup while preserving target-model accuracy, yielding substantial efficiency gains without compromising reasoning quality. |
| title | DREAM-R: Multimodal Speculative Reasoning with RL-Based Refined Drafting, Precise Verification, and Fully Parallel Execution |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2605.28678 |